What About Liquid Hydrogen Or Hydrogen Pipelines For Truck Stops? Yeah, No.

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

More publications on hydrogen refueling of freight trucks means more questions to answer. Freight trucking researcher, Jakob Rogstadius, Swedish senior researcher at RISE for Mobility and Systems, was working to understand the critique of The International Council on Clean Transportation’s fatally flawed total cost of ownership study and provided the English-language version of the German working group paper on hydrogen vs electric trucking costs. My assessment of the latter triggered more cogent questions, and I responded. Lightly edited, here’s what I wrote to him.

One point in the assessment caught Rogstadius’ eye, that 14 tanker trailers would be required to deliver the same energy in the form of compressed hydrogen to trucking stations as a tanker of diesel. Rogstadius asked:

“Regarding tanker trucks only being able to carry a puff of hydrogen, a colleague sent me this claim of 4.5 tonnes of liquid H2 per truck. I am not able to turn that information into an estimate of the impact on distribution costs.”

Liquid H2 must be chilled to 20° Kelvin through multiple chilling and thermal management cycles. That’s -253° Celsius or -424° Fahrenheit. Boil off rate in small tubular containers will be high.

Energy costs to liquify hydrogen are in the theoretical best possible case 10% of the energy value of the hydrogen and another 10% for the spin isomer conversion, one of the weird aspects of the physics of hydrogen. The actual efficiency in practice is that about a third of the energy in the hydrogen is used for liquification for transportation and storage. These are hard physics limits, not subject to radical improvements, but only slight incremental changes.

Liquid H2 experiences boil off at high rates. That’s the process by which the liquid with thermal inputs like the temperatures humans can survive at converts to gaseous hydrogen. Liquid natural gas tanker ships with huge, highly insulated, globular tanks and much higher temperatures than liquid hydrogen, experience boil off losses of 0.15% per day per regulations, and a bit higher in practice. There’s a reason why LNG ships are the only ships on the sea running on methane burning engines, as they’ve turned an issue into a feature, and I’m all for LNG ships continuing to do that as long as LNG ships last, because otherwise it would be vented to the atmosphere.

Paul Martin and I expect boil off rates for liquid hydrogen tankers of equivalent scale to LNG ships with be 0.5% or more per day in operation. For any smaller vessels, non-globular vessels containing liquid hydrogen, especially ones where weight is an issue and hence there will be an insulation/weight tradeoff, boil off will be higher. I would expect 2% to 5% losses from boil off at truck filling, truck movement and truck transfer of hydrogen to the refueling station as a reasonable rule of thumb. Note that gaseous hydrogen is a fairly potent greenhouse gas in its own right, so the vented hydrogen would be required to be captured and compressed or re-liquified, something that’s entirely doable, but of course adds more complexity and expense.

For the same boil off reasons, it’s unlikely that hydrogen would be stored in liquid form at refueling stations, but instead in much cheaper, larger, pressurized tanks in gaseous form. Transferring liquid hydrogen at 20° Kelvin from a tanker truck to a gaseous storage unit is a careful dance of massive input of heat to prevent the gaseous tank systems from freezing. Boil off hydrogen would have to be reliquified at the station that was using liquid hydrogen storage or there would have to be additional compression and gaseous hydrogen storage. More cost and complexity.

And hydrogen is a weird gas, one of three that in some circumstances actually gets warmer when decompressing, so there are steps in the chain where heat has to be removed. Once again, all doable, but non-trivial.

End to end, you lose a lot of the energy in hydrogen through liquification and transportation. With labor and capital costs for all the bits and bobs, it ends up costing a lot, just like trucking compressed gaseous hydrogen. Personally, I wouldn’t want to be near either a liquid hydrogen or 700 bar compressed hydrogen truck during a collision. Both would see explosive expansion and the liquid hydrogen would of course flash freeze everything and everyone it touches in the process. We transport hydrogen both ways today only for places where it’s critical to have hydrogen at a place we can’t manufacture it directly, like NASA.

Another point of the piece on the German study was my suggestion that pipelines of hydrogen to refueling stations were deeply unlikely. Rogstadius reasonably questioned that assertion as well.

“I believe the idea is to have a gas network and relatively few fueling stations, not to have distribution as ubiquitous as charging opportunities. On what basis do you conclude that “Plumbing hydrogen pipelines to refueling stations is deeply unlikely as well“?”

Don’t forget that truckers refuel when they need regulated breaks. The fewer the refueling stations, the more loss of movement time as truckers pull over into layoffs for mandated rests and then stop for refueling and eating as a separate step. There’s an optimal number in there, but it’s not a tiny one in a major geography with a network of roads.

As a note, achieving the 700 bar compression at the pump to the truck limits the rate of flow of hydrogen. The best achieved number in purely prototype NREL facilities is at the rate of diesel, but what that will look like in commercial operational conditions and at commercially justifiable costs of equipment is still anyone’s guess. Experience globally is that pumping hydrogen is a slower than pumping diesel or gasoline, and even then, a frequent failure condition is the thermal conditions causing icing of the pump to the vehicle requiring 5-10 minute waits for it to thaw by itself.

Assuming that a sparse network of much less used by every road traveler refueling stations with hence many fewer amenities and revenue streams will be satisfactory for anyone or cheaper seems to be a defect of logic on the part of those advocating hydrogen for trucking. I suspect it’s Toyota Mirai drivers with Stockholm Syndrome, who have converted their experiences hunting down the rare hydrogen stations — Austria has seven compared to 22,000 charging points — which have no washrooms, diners or other amenities into a virtue, and think anyone will find that pleasurable.

In other words, hydrogen refueling has to be integrated with more existing truck stops or it doesn’t make much sense for long haul trucking.

Hydrogen pipelines have their own challenges of course. Embrittlement of metal at micro fracture points occurs with normal pressure changes during gas compression for shipping and frost heaves. Higher grade metals and thicker pipes are required. Due to the low density of the gas, three times more compression energy is required to move the same units of energy in the gas as for natural gas as a comparator.

The ICCT study had a very odd choice of a 1 MW electrolyzer creating 500 kg of hydrogen per day, and a couple of additional scenarios with up to 2,500 kg per day. For context, long haul trucks might carry 1,000 liters of diesel and hence would require in the range of 500 kilograms of hydrogen for the same distance traveled. They included a lot of smaller vehicles as well, and Toyota Mirai’s carry about 5 kg. 2,500 kg a day of hydrogen is a tiny amount for a pipeline and unless there’s a complete local hydrogen gas utility with an existing pipe network —which there won’t be as all residential and commercial heating and cooking will be electric — a one off hydrogen pipeline to a trucking station would be a fairly expensive infrastructure project in its own right. Hydrogen pipelines would require very high guaranteed volumes.

The way the discussion is trending, the only vehicles where there’s actually any real remaining discussion is for long-haul N3 — the European equivalent of North American Class 8 — trucks. All local drayage and distances under 500 km are clearly going battery electric because the economics and battery energy density are so clearly suitable today for that, so the hydrogen refueling infrastructure is going to be limited to longer haul N3 trucks as potential customers. The ICCT TCO study does no one any services by having all classes of trucks in it as potential candidates for hydrogen, but that’s part and parcel of their focus on trying to fit the square peg of hydrogen into the round hole of ground transportation.

The combination doesn’t lead to hydrogen pipelines to truck stops seeming viable to me. For that to be viable, hydrogen would have to be used regularly in home heating and cooking, industrial heating, in other energy applications and in more modes of transportation. And it won’t be, despite current European and especially German illusions that it will be.